In its most common form, GIC is a phenomenon that takes place when high magnetic fields produced by electric particles emanated from the sun, periodically impinging our planet interact with the conductors of transmission and distribution circuits. Such interaction causes, according to the laws of physics, the induction of currents in these circuits. GIC can therefore flow in the network reaching the power transformers as well as the instrument transformers, shunt reactors and phase shifters connected to the transmission lines, circulating through their phase connections into their earthed neutral;. The most important effects are related to the saturation of those apparatus' magnetic circuitry. In general it can cause wave distortion and equipment overheating. Possible outcomes of this disturbance are the malfunction of protective systems and/or failure as well as a deterioration of the grid's performance, including voltage collapse.
The last well-remembered event that epitomizes the potential harmful consequences of this phenomenon; took place in March 1989, when a geomagnetic storm produced a major blackout in eastern Canada and part of the northeastern US, leaving the complete region powerless for many hours, affecting several million customers. Moreover the same meteor caused the destruction of a nuclear-plant's generator transformer in eastern United States.
Three basic approaches have been suggested in the literature to cope with this phenomenon; they are discussed as follows:                1) re-engineering existing power apparatus prone to the GIC effects, so that they perform more robustly under these conditions. U.S. Pat. No. 5,179,489 (1993) proposes altering the magnetic circuits of transformers in the system to reduce half-cycle saturation and associated high current peak; also providing means of countering the difference of potential between spaced grounding points in the system;        2) turning the power grid off causing a planned blackout when a magnetic storm is imminent; Molinski discusses this idea in an article of the IEEE Spectrum Magazine (2000);        3) adding active blocking components to the transformer to impede GIC flow, as per U.S. Pat. No. 5,751,530 (1998), proposing a capacitor device from the neutral of the transformer to ground. It is noted that this document makes a description of the problem and circuitry involved. U.S. Pat. No. 5,436,786 (1995) blocks GIC by means of controlled SCRs with surge protection. Also U.S. Pat. No. 6,067,217 (2000) even though unrelated to GIC, proposes current limiting hardware of the active type to limit fault currents, in this latter case not from neutral to ground but in each of the three phases instead.        
Disadvantages of these approaches:
While 1) would potentially improve the performance of the apparatus, its planned-outage timing and cost, engineering challenges, added loss of revenue and system reliability during such a process makes it absolutely unrealistic.
Regarding 2), while it will definitely protect the integrity the system, blackouts are very onerous and dicey. However is questionable that it can be done effectively due to the short lead time from the storm front detection (less than an hour); most thermal generating plants require one hour o more to shut down. Nevertheless it is not possible to predict ahead of time whether the intensity of a storm will warrant the extremeness of a self-inflicted blackout.
Finally 3) appears as an obvious way to knock this current, yet quite compromising in this application; for in these cases transient-phenomena wave distortions developed under GIC, the presence of capacitors in series with the transformer, can cause undesired ferroresonant conditions for some of the numerous natural frequencies created. Moreover the required capacitor must be of an impractical large size to minimize its steady-state impedance impact. In addition switching capacitors on/off is never trouble free; neither discharging them is. But in this case both things are required. Indeed this becomes a risky proposition stemming from the fact that the full charge induced in the transmission line/transformer by the magnetic storm is stored in this very neutral capacitor. For this element is sitting in a critical transformer location; therefore this phenomenology requires active components to reckon with. Consequently the proposed design includes a number of electronic/control components which pose reliability concerns of their own, not to mention cost-effectiveness. In conclusion this concept, which implies a problematic prosthesis to existing equipment, would most certainly be considered unacceptable for the utility industry as it is also suggested in the aforementioned IEEE Spectrum article.
Likewise considerations entail the addition of different components to the phases of transformers, i.e. in a phase-to-neutral mode as opposed to neutral-to-ground, as will be presented below. This idea is very dependent on controlled switching, which includes plenty power electronics; it can be stated again that it becomes a major and compromising prosthesis to existing equipment; it may just compound (triplicate) some of the reliability problems described above. Furthermore incorporating hardware in the respective phases imposes both a fill current/power rating and three-phase fault duty. On the other hand the performance of an active-type means, requiring command-and-control, would be considered very questionable under magnetic storms which normally jam these systems.
In sum, it should be acknowledged that the inability on the part of the industry to adopt an acceptable GIC countermeasure is a serious matter, for it leaves it unprotected before future solar storms while it has been established that some of them could potentially be very damaging to the electric power infrastructure.